Please find the new patch in http://marc.info/?l=linux-spi&m=143816718611912&w=2 Thanks, Sifan > -----Original Message----- > From: Sifan Naeem > Sent: 27 July 2015 17:54 > To: 'Mark Brown'; Andrew Bresticker > Cc: linux-spi@xxxxxxxxxxxxxxx; Stable kernel (v4.1) > Subject: RE: [PATCH 2/2] spi: img-spfi: fix spfi_setup by removing > gpio_request_one > > Thanks Mark, Andrew. > > I'll rework this and submit a new patch. > > Thanks, > Sifan > > > -----Original Message----- > > From: Mark Brown [mailto:broonie@xxxxxxxxxx] > > Sent: 27 July 2015 17:40 > > To: Andrew Bresticker > > Cc: Sifan Naeem; linux-spi@xxxxxxxxxxxxxxx; Stable kernel (v4.1) > > Subject: Re: [PATCH 2/2] spi: img-spfi: fix spfi_setup by removing > > gpio_request_one > > > > On Mon, Jul 27, 2015 at 09:34:41AM -0700, Andrew Bresticker wrote: > > > On Mon, Jul 27, 2015 at 5:11 AM, Sifan Naeem > > > <sifan.naeem@xxxxxxxxxx> > > wrote: > > > > > > spfi_setup may be called many times bye the spi framework, but > > > > gpio_request_one can only be called once without freeing, > > > > repeatedly calling gpio_request_one will cause an error to be > > > > thrown, which causes the request to spi_setup to be marked as failed. > > > > > > We can simply use gpio_direction_output to set the direction of > > > > the gpio instead of gpio_request_one to put the gpio in to initial > > > > state, after which the spi framework can control the chipselect > > > > line via gpio using gpio_set_value. > > > > > I don't think we want to leave the CS GPIOs un-requested. Instead, > > > we could either request them all at probe() time (and then set > > > direction in setup()) or have a per-spi_device flag that indicates > > > whether or not the GPIO has been requested. > > > > Not requesting the GPIOs at all would just be a straight up bug. -- To unsubscribe from this list: send the line "unsubscribe linux-spi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html